New cost function for motion estimation in MPEG encoding using LNS

The Logarithmic Number System (LNS) has lower power and larger dynamic range than fixed point, which makes LNS suitable for designing low-power, portable devices. Motion estimation is a key part of the MPEG encoding system. This paper introduces LNS into motion estimation for the MPEG encoding system. The block matching technique is the most commonly used motion-estimation method in MPEG encoding. The Mean Absolute Difference (MAD) is an inexpensive fixed-point cost function, which uses the sum of the absolute difference of the pixel values in the reference and encoded frames. Since LNS addition and subtraction are expensive, we propose the quotient of the two pixels' values instead of the difference. LNS division only needs a fixed-point subtractor. Similar to the absolute difference, we take the quotient of the larger value over the smaller value. We call this new cost function Mean Larger Ratio (MLR). The product of such ratios is calculated for each of the macroblocks in MPEG frames. Using MLR, LNS has approximately the same hardware as MAD for fixed point. Example videos show MLR provides a practical cost function to perform motion estimation with LNS.

[1]  Vassilis Paliouras,et al.  Low-power properties of the logarithmic number system , 2001, Proceedings 15th IEEE Symposium on Computer Arithmetic. ARITH-15 2001.

[2]  Tihao Chiang,et al.  A novel all-binary motion estimation (ABME) with optimized hardware architectures , 2002, IEEE Trans. Circuits Syst. Video Technol..

[3]  Yuk Ying Chung,et al.  Custom computing implementation of two-step block matching search algorithm , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[4]  Sung-Jea Ko,et al.  New motion estimation algorithm based on bit-plane matching and its VLSI implementation , 1999, Proceedings of IEEE. IEEE Region 10 Conference. TENCON 99. 'Multimedia Technology for Asia-Pacific Information Infrastructure' (Cat. No.99CH37030).

[5]  Heung-Kyu Lee,et al.  An efficient block-matching criterion for motion estimation and its VLSI implementation , 1996 .

[6]  Mark G. Arnold Reduced power consumption for MPEG decoding with LNS , 2002, Proceedings IEEE International Conference on Application- Specific Systems, Architectures, and Processors.

[7]  Jie Ruan,et al.  Threshold Mean Larger Ratio Motion Estimation in MPEG Encoding Using LNS , 2004, PATMOS.

[8]  Borko Furht,et al.  Motion estimation algorithms for video compression , 1996 .

[9]  Mark G. Arnold LNS for low-power MPEG decoding , 2002, SPIE Optics + Photonics.

[10]  N. Kingsbury,et al.  Digital filtering using logarithmic arithmetic , 1971 .

[11]  Hsueh-Ming Hang,et al.  A comparison of block-matching algorithms mapped to systolic-array implementation , 1997, IEEE Trans. Circuits Syst. Video Technol..

[12]  Lawrence A. Rowe,et al.  Analysis Tools for MPEG-1 Video Streams , 1997 .

[13]  Israel Koren Computer arithmetic algorithms , 1993 .

[14]  Jie Ruan,et al.  Combined LNS Adder/Subtractors for DCT Hardware , 2003, ESTImedia.